A new membrane design could change the way scientists think about separating lithium, sodium and potassium ions. Researchers have developed a covalent organic framework monolayer that uses the water shells surrounding ions as an active part of the separation process, rather than forcing ions through rigid pores by stripping away most of their hydration. The result is an angstrom-scale “hydrapore” that can distinguish between monovalent cations while allowing them to move at unusually high speed. In tests performed under a 1-molar concentration gradient, the membrane achieved a selectivity of 148 for potassium over lithium and 42 for sodium over lithium, while maintaining a potassium permeance of 2 × 10⁴ mol m⁻² h⁻¹. According to the researchers, this combination of precision and throughput is far beyond that of many existing ion-separation membranes.
The work addresses a fundamental problem in nanofluidic separation. Ions dissolved in water are not bare charged particles. Each ion attracts surrounding water molecules, creating a dynamic hydration shell that moves with it. The strength and structure of this shell depend on the ion’s charge density, size and interactions with water. Lithium ions, for example, bind water particularly strongly and carry a comparatively stable hydration environment. Potassium ions are larger and generally less tightly hydrated, while sodium occupies an intermediate position. Conventional nanopores often exploit differences in hydrated or dehydrated ion size, but they typically require ions to shed part of their water shell before entering a narrow channel. That process consumes energy and can dramatically slow transport.
Rigid nanopores therefore face a difficult trade-off. Making the pore smaller can improve selectivity, but it also increases the energetic barrier that ions must overcome. A pore that is too narrow may force substantial dehydration, creating friction and resistance as the ion passes through. A pore that is slightly wider may allow faster movement, but ions with different properties can then travel through with insufficient discrimination. This tension has limited the ability of membranes to combine high selectivity with high permeability, particularly for monovalent ions such as lithium, sodium and potassium, which all carry the same single positive charge.
The new strategy turns hydration from an obstacle into a molecular control mechanism. The researchers anchored hydrated ions at the rim of pores in a covalent organic framework, a porous, crystalline material assembled through strong chemical bonds. These anchored ions do not simply block the openings. Instead, their hydration layers create a dynamic environment that interacts with incoming ions. As a migrating ion approaches, the hydration shells around the bound and mobile ions can merge, compress and rearrange. Those changes generate attractive or repulsive forces that regulate whether the ion can enter and how easily it can pass through the pore.
This mechanism differs from conventional size-exclusion sieving. In a standard molecular sieve, the pore is treated as a mostly fixed opening, and transport depends largely on whether a particle can physically fit. In the new membrane, the effective pore is not defined only by the framework’s geometric dimensions. It is also shaped by the hydration layers at the pore edge. These water-rich regions act like a soft, responsive gate at angstrom-scale dimensions. Because the gate can interact differently with different hydrated ions, the membrane can discriminate between ions without requiring complete dehydration.
The distinction between potassium, sodium and lithium arises from the way each ion interacts with water and with the hydrated species positioned at the pore rim. When a migrating ion enters the hydrapore, the degree to which its hydration layer merges with or compresses the anchored hydration layer influences the energetic cost of transport. An ion that produces a favorable interaction may be guided through the opening, while another may experience repulsion or a larger barrier. In this way, the membrane uses the physical chemistry of hydration shells to separate ions that would otherwise be difficult to distinguish because they share the same charge.
The reported performance is striking because it combines a high degree of selectivity with rapid ion flow. Under a concentration difference of 1 molar, the membrane showed a potassium-to-lithium selectivity of 148. This means potassium transport was strongly favored over lithium transport under the reported testing conditions. The sodium-to-lithium selectivity reached 42, also indicating substantial discrimination. At the same time, potassium permeance reached 2 × 10⁴ mol m⁻² h⁻¹. Permeance describes how much material crosses a membrane per unit area and time under a driving force; the value reported here is approximately three orders of magnitude higher than those associated with state-of-the-art membranes cited by the researchers.
The membrane’s low activation energy provides another clue to why transport is so fast. The measured value was 5.5 kilocalories per mole, suggesting that ions encounter only a relatively small energetic barrier as they move through the hydrapores. In practical terms, the ions are not being dragged through a constricted, dehydrating channel. Instead, their hydration environments appear to reorganize in a way that permits movement with limited friction. The researchers describe this behavior as near-frictionless transport, although the phrase refers to the unusually low transport barrier at the nanoscale rather than the complete absence of resistance.
The concept could be especially important for technologies that depend on selective monovalent-ion separation. Lithium recovery, desalination, electrochemical energy storage and advanced water treatment all require membranes that can distinguish closely related ions without excessive energy consumption. Existing approaches may lose efficiency when ions must repeatedly dehydrate and rehydrate, or when high selectivity is achieved only by sacrificing flux. A hydration-layer-mediated membrane offers a different design principle: instead of making the pore increasingly rigid and narrow, engineers could tune the chemical identity, position and hydration behavior of ions or functional groups around the opening.
The study also points toward a broader change in membrane science. For decades, researchers have focused on controlling pore size, surface charge and chemical affinity. Those parameters remain important, but the new findings show that the surrounding water structure can be engineered as well. Hydration layers are constantly changing, yet they can still provide a reliable basis for molecular recognition when confined at a pore rim. The approach may inspire membranes whose selectivity comes from coordinated interactions among the framework, bound ions, water molecules and migrating ions. If the performance can be reproduced at larger scales and maintained in complex mixtures, dynamic hydrapores could offer a route toward faster, more energy-efficient ion separations than rigid nanopores allow.
Subject of Research: Hydration-layer-mediated separation of monovalent cations using dynamic angstrom-scale hydrapores in a covalent organic framework monolayer.
Article Title: Hydration layer mediates near-frictionless sieving of monovalent ions
Article References: Yang, J., Tu, B., Liu, Y. et al. “Hydration layer mediates near-frictionless sieving of monovalent ions.” Nature Nanotechnology (2026). https://doi.org/10.1038/s41565-026-02245-7
Image Credits: AI Generated
DOI: https://doi.org/10.1038/s41565-026-02245-7
Keywords: ion separation, monovalent cations, hydration layers, hydrapores, covalent organic frameworks, nanoporous membranes, lithium separation, potassium transport, sodium separation, nanofluidics
